ORIGINAL ARTICLE

Peer Toy Play as a Gateway to Children’s Gender Flexibility: The Effectof (Counter)Stereotypic Portrayals of Peers in Children’s Magazines

Lauren Spinner1 & Lindsey Cameron1& Rachel Calogero2

# The Author(s) 2018. This article is an open access publication

AbstractExtensive evidence has documented the gender stereotypic content of children’s media, and media is recognized as an importantsocializing agent for young children. Yet, the precise impact of children’s media on the endorsement of gender-typed attitudes andbehaviors has received less scholarly attention. We investigated the impact of stereotypic and counter-stereotypic peers picturedin children’s magazines on children’s gender flexibility around toy play and preferences, playmate choice, and social exclusionbehavior (n = 82, age 4–7 years-old). British children were randomly assigned to view a picture of a peer-age boy and girl in amagazine playing with either a gender stereotypic or counter-stereotypic toy. In the stereotypic condition, the pictured girl wasshown with a toy pony and the pictured boy was shown with a toy car; these toys were reversed in the counter-stereotypiccondition. Results revealed significantly greater gender flexibility around toy play and playmate choices among children in thecounter-stereotypic condition compared to the stereotypic condition, and boys in the stereotypic condition were more acceptingof gender-based exclusion than were girls. However, there was no difference in children’s own toy preferences between thestereotypic and counter-stereotypic condition, with children preferring more gender-typed toys overall. Implications of thefindings for media, education, and parenting practices are discussed, and the potential for counter-stereotypic media portrayalsof toy play to shape the gender socialization of young children is explored.

Keywords Early childhood development . Stereotyped behavior . Gender role attitudes . Gender flexibility . Toy play . Mediaexposure . Children’s print magazines . Social acceptance . Bullying

Gender-normative attitudes and behaviors, and their accom-panying stereotypes, dominate children’s media and popularculture (Blakemore and Centers 2005; Leaper et al. 2002;Murnen et al. 2016; Thompson and Zerbinos 1995).Portrayals of boys tend to emphasize masculine gender rolesand stereotypically masculine play and toys, whereas por-trayals of girls tend to emphasize feminine gender roles andstereotypically feminine play and toys (Cherney and London2006; Kahlenberg and Hein 2010). These gendered messagesare communicated through various forms of children’s media,

including television programming and advertisements (Bakir2013; Bakir and Palan 2013; Merskin 2002), books (Foster2016; Skinner 2013), video games (Miller and Summers2007; Sheldon 2004), and print magazines (Spinner et al.2016).

Exposure to gender-stereotypedmodels in children’s mediahas implications for children’s social and gender-specific de-velopment (Coyne et al. 2014; Signorielli 2001). One impor-tant domain that has been understudied is the impact of peerson children’s gender flexibility in their preferences for toysand playmates. In the present study, we build on previousinvestigations of the impact of gendered media on childrenby testing the effect of exposure to gender-typed toy play bypeers pictured in children’s print magazines on gender flexi-bility in toy and playmate preferences in young children. Inparticular, we examined the extent to which various indicatorsof children’s gender flexibility, including gender-based socialexclusion, may be undermined and/or bolstered by peers’(counter)stereotypic displays of toy play through this medium.

* Lauren SpinnerL.Spinner@kent.ac.uk

1 School of Psychology, University of Kent, Canterbury, Kent CT27NP, UK

2 Department of Psychology, University of Western Ontario,London, ON, Canada

Sex Roleshttps://doi.org/10.1007/s11199-017-0883-3

Gender Flexible Attitudes and Behavior

Gender flexibility refers to an open-minded attitude aroundgender roles. Ruble and Martin (1998, p. 947) defined it asBthe willingness to apply an attribute to both sexes, rather thanjust to one or the other, or the recognition of the relativity ofstereotypes (e.g., that norms could be different in anotherculture).^ Previous research has shown that gender flexibilityis acquired once gender-related knowledge has beenestablished (Huston 1983). So although children’s gender ste-reotype knowledge rapidly increases between the ages of 3and 6 years-old (Aubry et al. 1999), their acceptance of thesestereotypes as Bcorrect^ or Bfixed^ begins to decline, withgender flexibility peaking and then plateauing at around7 years of age, following a period of gender stereotype rigiditybetween the ages of 5 and 6 years (Miller et al. 2006; Huston1983; Signorella et al. 1993; Trautner 1992).

This trajectory of gender flexibility has been demonstratedby Halim et al. (2014) in their research on gender appearancerigidity among children aged 3–7 years of age. They foundthat younger girls were more motivated to dress in gender-typed ways than older girls were, and understanding of genderstability (i.e. knowledge that gender remains stable over time)predicted appearance rigidity in both boys and girls. There isalso evidence that children’s gender-typed play increases inrigidity between the ages of 3 and 4 years, remaining stableuntil age 5 (Halim et al. 2013). Similarly, when examining therole of gender development in Halloween costume choicesamong infants and preschoolers, Dinella (2017) found stronggendered trends in these costumes, with princess costumesbeing most popular for girls and superhero costumes for boysand with age being positively related to the gender-typing ofchildren’s costumes in this young sample. Together, thesefindings reflect a strengthening in gender-typed behavioramong pre-school children, as well as the emergence of gen-der flexibility among older children as they approach 7 yearsof age, as predicted by cognitive developmental theories ofgender development.

Flexibility around gender can be expressed in a multi-tude of ways and directed toward oneself and/or others,with children tending to show more tolerance towardothers’ gender-flexible behavior, but less so toward theirown gender-flexible behavior (Katz and Ksansnak 1994).Two specific contexts within which children might be ableto express gender-flexible behavior include their toy pref-erences and playmate preferences. Preferences for gender-typed toys and same-gender playmates begin to emergearound 2 years of age (Caldera et al. 1989; Maccoby andJacklin 1987; Serbin et al. 1994; Wood et al. 2002). Theentrenchment of gender stereotypes and prejudice at suchan early and formative stage of development has implica-tions for children’s identities, aspirations, and achieve-ments (Cimpian et al. 2012) as well as the perpetration of

gender-related bullying, peer victimization, and social ex-clusion (Killen and Stangor 2001).

Moreover, in westernized societies, gender segregation re-mains a salient feature of many people’s everyday workingand social lives, and it contributes to poor gender relationships(Leaper 1994). Gender segregation of peer groups is one of themost salient aspects of children’s social and cognitive devel-opment (Geary and Bjorklund 2000; Killen and Stangor 2001;Maccoby 2002). By 6 years of age, children spend significant-ly more time playing with children of the same gender com-pared to the other gender (Maccoby and Jacklin 1987), whichcan increase gender-typed behavior (Martin and Fabes 2001).In order to maintain gender-segregated peer groups and divi-sion among playmates, social exclusion may be necessary.Social exclusion can have severe consequences for children,including reduced academic motivation and success, and anegative impact on mental health and well-being (Buhs et al.2006). Identifying strategies to encourage mixed-gender andcounter-stereotypic play is useful because these experiencesexpose children to a wider variety of play styles and expandopportunities for cognitive and social development (Fabeset al. 2003). It is therefore important to find ways to encouragemixed-gender friendships in children as a means of attenuat-ing gender-typed behavior. We focus on gender flexibility inthe present study as one potentially malleable social-cognitivefactor that might improve gender relationships for childrennow as well as the adults they will later become.

Children’s Print Magazines as a GenderSocializing Agent

There are a number of theoretical accounts for how gender-related attitudes and behaviors develop and why they arerelatively inflexible. According to gender schema theory(Bem 1981, 1983), deeply rooted gender polarization incultural discourse and social institutions promotes the de-velopment of gender-based cognitive schemas in childrenat an early age whereby children acquire a learned readi-ness to evaluate, organize, and filter information and be-havior in terms of what boys and girls should and shouldnot do (Martin and Ruble 2004). From the perspective ofcognitive social learning theorists (Bussey and Bandura1999, 2004), environmental agents provide and rewardmodels of gender-normative behavior for children to ob-serve and imitate, thereby shaping and reinforcing gender-role attitudes and behavior. Cultivation theory argues thatthe repetition of themes and stereotypes over time in themedia, and television programming specifically, leadsviewers to cultivate beliefs about the real world that matchwith the media content (Gerbner 1998). Together, thesetheoretical accounts converge on the idea that male andfemale children are transformed into masculine and

Sex Roles

feminine adults through a variety of gender socializationforces and processes.

Media represent a powerful socializing agent of gender-role norms because they communicate our cultural definitionsof gender normativity in a myriad of formats and settings. Todate, much of the research on the impact of gender-stereotypicportrayals in media has been conducted in industrialized west-ernized societies (Collins 2011). Indeed, despite shifts in thegender roles assumed by women and men in recent decades(Rich 2005), as well as the increased professional achieve-ments of women (Hunt 2004), the United Kingdom, for in-stance, largely remains a Bmasculine^ society (Hofstede2001). Despite the fact that the gender pay gap is the lowestit has even been in the United Kingdom, women still earnmore than 18% less than their male counterparts, and occupa-tions remain highly gender-segregated (GovernmentEqualities Office 2016). The dominant portrayals of womenin popular British print magazine advertisements continue toperpetuate gender-stereotyped representations of them(Plakoyiannaki and Zotos 2009). Moreover, the actual andaspirational choices for occupations among young womenliving in the United Kingdom (n = 506; aged 13–18 years-old) continue to reflect deeply entrenched gender roles(Gould 2008), with markedly more female adolescents indi-cating they want to be models (32%) or actors (29%) com-pared to engineers (4%) or scientists (14%). This sexist cul-tural context provides an important site for investigation of theimpact of gender-stereotypic portrayals in children’s mediaand how we might attenuate it.

Children’s magazines represent a print-based medium thatremains popular among young children, with approximately1.8 million children’s magazines being sold in the UnitedKingdom in 2015 (Statista 2016a) and 45% of 5- to 7-year-olds in the United Kingdom being classified as regular readersof magazines, books, or comics (Statista 2016b). Children’smagazines present gender stereotypes through the images, ac-tivities, emotions, colors, advertisements, and narratives fea-tured in the pages. A unique feature of children’s magazines isthe use of reader’s pages, which feature photos of actualreaders of the magazine and information about them, asopposed to fictional and/or less identifiable characters.Shutts et al. (2010) demonstrated that children prefer objectsand activities endorsed by models of the same gender and ageas themselves, even though children fail to acknowledge theinfluence of these social categories on their decisions. We alsoknow that peers are strong enforcers of gender-normative play(Kornienko et al. 2016). We propose that portrayals of age-matched peers who share an interest with readers through themagazine may serve as effective social models for the com-munication of gender-typed attitudes and behaviors in mediaaimed at young children, especially regarding gender-typedtoy play.

Toys as Socializing Cultural Products

Children’s toys represent influential cultural products that arestrongly gender-stereotyped (Cherney and London 2006),even in societies with an explicit emphasis on gender equalitypolicies, such as Sweden (Nelson 2005). This pattern is un-surprising given the extent to which many popular toys featuregender-stereotyped characteristics in their design (Blakemoreand Centers 2005; Murnen et al. 2016) and are explicitly la-beled as Bboy toys^ or Bgirl toys^ in the marketing of theseproducts and within the stores where they are sold (Auster andMansbach 2012; Kahlenberg and Hein 2010)—consumerswould be hard-pressed to miss the Bpink aisle^ (targetinggirls) in any major toy store. Findings from experimental stud-ies indicate that children prefer gender-typed toys in terms ofboth their function and color (Weisgram et al. 2014;Wong andHines 2015). For example, Weisgram et al. (2014) found thatboys prefer masculine to feminine toys and that girls dislikemasculine toy and color combinations more than any other toytype and color combination. Research has also shown thatchildren’s toy preferences are influenced by the way in whichtoys are modeled and who is modeling them (Bradbard andEndsley 1983). Children favor novel toys when they are iden-tified with the children’s own gender, and toys modeled by asame-gender child are rated as more attractive (Shell andEisenberg 1990).

This gender divide in toy preferences merits scientificand practitioner interest because different types of toysfacilitate different types of play, and play types have beenassociated with different developmental trajectories for so-cial and cognitive skills in children. Research with youngchildren has shown that traditional toys for boys (e.g., cars,video games) facilitate the development of visuo-spatialskills and promote a more agentic orientation toward selfand others (De Lisi and Wolford 2002; Jirout andNewcombe 2015), whereas traditional toys for girls (e.g.,dolls, Disney princesses) facilitate the development of nur-turing and empathy skills and promote a more communaland appearance-focused orientation toward self and others(Coyne et al. 2016; Dittmar et al. 2006; Li and Wong2016). In addition, there is evidence that children’s culturalproducts, including toys, are becoming more sexualized ingender-divergent ways (Boyd and Murnen 2017;Zurbriggen and Roberts 2013). One study has also linkedgender-stereotyped toy play to the career cognitions of 4–7 year-old children (Sherman and Zurbriggen 2014).Specifically, girls who played with Barbie indicated fewerfuture career options for themselves compared to what theyindicated for boys, whereas girls who played with Mrs.Potato Head did not indicate such differences in futurecareer options. Thus, the toys with which children preferto play matters for their overall development.

Sex Roles

Counter-Stereotypic Modelsand Gender-Flexibility

Research findings suggest that gender-typed toy preferencesand attitudes are malleable and can change in response toexposure to gender counter-stereotypic models (Abad andPruden 2013; Steyer 2014). Indeed, if stereotypic portrayalsand models provide one mode of gender socialization, thencounter-stereotypic portrayals and models provide anothermode of gender socialization. For example, after a brief expo-sure to counter-stereotypic portrayals of women in televisioncommercials (vs. stereotypic portrayals), both girls and boysreported less gender-typed views toward women (Pingree1978). After one exposure to counter-stereotypic (vs. stereo-typic) portrayals of female characters in children’s books, bothgirls and boys demonstrated stronger preferences for gender-neutral toys compared to gender-typed toys (Ashton 1983).However, the strength of the effect of counter-stereotypicmodels on these preferences and attitudes may be moderatedby children’s own gender.

For instance, focusing specifically on highly gender-typedchildren over a 4-month period, researchers demonstrated asignificant shift away from gender-typed toy play after expo-sure to fictional stories featuring gender-neutral and gendercounter-stereotypic toy play, but only for girls (Green et al.2004). Pike and Jennings (2005) further demonstrated thatyoung participants exposed to 3 min of video footagedepicting Breal children^ engaged in toy play in traditional(all boys) or nontraditional (all girls) settings in televisioncommercials were more likely to categorize toys as appropri-ate for Bboth boys and girls^ if they have seen the nontradi-tional commercial, and this effect was stronger for boys thanfor girls. Additionally, research has shown that boys are morelikely to imitate same-gender models than girls are, and boyshave been found to be particularly reluctant to imitate femalemodels or male models if they are performing gender-atypicalbehaviors (Bauer 1993; Perry and Bussey 1979; Slaby andFrey 1975).

Adherence to gender-typed toy play has been found to beparticularly strong among boys, because boys are discouragedfrom play aligned with feminine stereotypes whereas girls areencouraged to play in masculine-typed ways to raise theirstatus (Cahill and Adams 1997). In relation to gender-typedcolors, previous research has shown that although boys in-creasingly avoid pink during the early years of development,there is no evidence to show that girls avoid blue (LoBue andDeLoache 2011). Similarly, although pre-school children havebeen shown to categorize occupations in line with genderstereotypes (Blakemore 2003; Liben et al. 2002), young chil-dren often permit women to occupy masculine-typed occupa-tions, but do not permit men to occupy feminine-typed occu-pations (Schuette et al. 2012).

Encounters with counter-stereotypic gender-related behav-ior may also impact gender-related attitudes and behavior be-yond toy play preferences. Research has demonstrated thatself-perceptions, interests, and pursuits are affected by expo-sure to gender counter-stereotypic models. For example, 1113rd and 4th grade boys and girls exposed over a 4-week periodto female protagonists in children’s books who displayedgender-atypical behavior increased the number of activitiesand occupations they identified as gender-appropriate forwomen to undertake (Scott and Feldman-Summers 1979).Children who were assigned gender-neutral textbooks to prac-tice their reading later judgedmore activities as appropriate forgirls and boys than those who were assigned gender-typedtextbooks (Karniol and Gal-Disegni 2009). Nhundu (2007)also found that girls in Zimbabwe who read biographicalstories of women in gender atypical careers adjusted theirown career aspirations in non-traditional directions. Overall,given the fact that pervasive portrayals of gender stereotypesmore broadly serve to channel and limit children’s interests,experiences, and activities over time (Serbin et al. 1994), theseresearch findings underscore the importance of investigatingthe potential for counter-stereotypic models and representa-tions of gender-related behavior to increase children’s genderflexibility.

Less research has examined the effects of counter-stereotypic gender portrayals on children’s perceptions of oth-er children and their behavior toward them. In one relevantstudy, using the Playmate and Play Style PreferencesStructured Interview (PPPSI) and cartoon depictions of peers,Pasterski et al. (2011) presented children with a social conflictwhereby they had to choose between an other-gender play-mate who was playing with a same-gender toy (e.g., for boys,a girl playing with vehicles) or a same-gender playmate whowas playing with an other-gender toy (e.g., for boys, a boyplaying with a tea set). They demonstrated that boys choseplaymates based on the play style of the peer rather than thepeer’s gender label, whereas girls chose playmates based onplay style and peer gender label. Thus, play style, rather thangender alone, may underlie gender-segregated play in chil-dren. These findings are consistent with research on thecognitive-behavioral similarity model, which proposes thatchildren can overcome preferences for same-gender peers ifthere are behavioral similarities with an other-gender peer(Martin et al. 2011). For instance, a boy may display a similarpreference for playing with a girl who enjoys trucks as hewould for playing with a boy. In other words, children whoengage in counter-stereotypic play may be integral to normal-izing gender desegregation and gender inclusion.

The overall findings from Pasterski et al.’s (2011) studysuggest that a perceived shared interest in a play activitymay be a critical piece for cultivating gender flexibility andreducing social exclusion because children’s preferences for

Sex Roles

gender-typed toys and toy play appear earlier in developmentand before the emergence of gender-segregated group play(Campbell et al. 2002; Serbin et al. 1994). To date, there islimited research on this possibility in young children.Research with older children has indicated that by the age of9-years-old, children are aware of the potential for exclusionby their peers if they challenge gender-stereotypic groupnorms by engaging in counter-stereotypic activities, especiallyif boys try to engage in female-stereotypic activities (Mulveyand Killen 2015). It is less desirable for boys to exhibit fem-inine behavior or engage in feminine activities than it is forgirls to exhibit masculine behavior or engage in masculineactivities, and therefore boys are more likely than girls are tobe penalized and excluded by peers for breaking from gendernorms regarding activity choices (Blakemore 2003; Horn2008). This pattern suggests that boys may be more likely tomake playmate choices based on toy-play, rather than on gen-der of playmate, whereas girls use gender and toy-play infor-mation when choosing their playmates.

The Present Study

The present study integrates and extends previous research onthe effects of gender stereotypic versus counter-stereotypicmedia portrayals of children on a set of gender-flexible atti-tudes and behavior in young British children. We focused onthe impact of portrayals of children engaged in gender-stereotypic or counter-stereotypic toy play in print magazines,depicted in the form of actual children playing with their toysand who were fellow readers of the magazine, that is, in aformat made to resemble the content of a Reader’s Page thatis often found in children’s magazines. The portrayals of thechildren included an age-matched male and female child tobolster the validity and potential impact of the peer (Bartini2006). The children were depicted as playing with a toydeemed appropriate for their own gender (stereotypic toyplay) or a toy deemed appropriate for the other gender (coun-ter-stereotypic toy play). This design allowed us to randomlyassign children to view (a) magazine content that pictured aboy and girl engaged in stereotypic toy play or (b) magazinecontent that pictured a boy and girl engaged in counter-stereotypic toy play.

We also used a variety of markers of gender flexibility toassess the degree to which the magazine content would differ-entially shift the gender-related preferences and attitudes ofyoung children. Specifically, we examined whether exposureto counter-stereotypic (vs. stereotypic) peers through this me-dium would impact preferences for gender-typed toys (seeHypotheses 1a and 1b), attitudes toward gender-typed toyplay (see Hypotheses 2a and 2b), playmate preferences (seeHypotheses 3 a–c), and the endorsement of gender-based so-cial exclusion (see Hypotheses 4a and 4b). The focus on

playmate preferences and gender-based social exclusion rep-resent particularly understudied outcomes among this devel-opmental age group of 4–7 year-olds, especially in the contextof stereotyped media content exposure. We focused on thisage range because it is between these ages that children’sgender identity and gender-related knowledge, attitudes, andbehaviors develop significantly (Serbin and Sprafkin 1986;Signorella et al. 1993; Zosuls et al. 2009).

For gender-typed toy preferences, we expected children tomake gender-typical toy preferences, as evidenced by an in-teraction between participant gender and toy type, wherebyboys would prefer to play with masculine toys over femininetoys and girls would prefer to play with feminine toys overmasculine toys (Hypothesis 1a). We also expected conditionto moderate children’s gender-typed toy preferences,predicting a three-way interaction among participant gender,condition, and toy type whereby children in the counter-stereotypic condition would prefer other gender toys morethan children in the stereotypic condition would, demonstrat-ing greater gender flexibility around toy type (Hypothesis 1b).

For attitudes toward gender-typed toy play, we expected amain effect of participant gender whereby girls would demon-strate more gender flexible attitudes toward toy play than boys(Hypothesis 2a). We also expected a main effect for conditionwhereby children in the counter-stereotypic condition, com-pared to children in the stereotypic condition, would be morelikely to label toys as being for both boys and girls, demon-strating more gender flexible attitudes around toy play(Hypothesis 2b).

For gender-typed playmate choice, we expected children todemonstrate more gender flexible attitudes around playmatepreferences in the counter-stereotypic condition comparedwith the stereotypic condition. We expected children to bemore likely to choose a same-gender than an other-genderplaymate in the stereotypic condition, whereas we did notexpect to observe this bias in the counter-stereotypic condition(Hypothesis 3a). Also in the counter-stereotypic condition, weexpected that boys would bemore likely than girls would be tochoose an other-gender playmate compared to a same-genderplaymate. This is because, compared to girls’, boys’ playmatepreference may be more driven by prospective playmates’ toychoice, rather than by their gender, due to more stronglyenforced norms for traditional masculine behavior(Hypothesis 3b). We also expected that the reasons childrenwould provide for their playmate choice would be more likelyto refer to toy play style than the playmate’s gender in thecounter-stereotypic condition, whereas we expected toy playstyle and playmate’s gender to be given as reasons in thestereotypic condition (Hypothesis 3c).

For gender-based social exclusion, we expected children todemonstrate more gender flexible attitudes in the counter-stereotypic versus stereotypic condition. We expected a maineffect for condition, whereby children in the counter-

Sex Roles

stereotypic condition would report less endorsement ofgender-based social exclusion than children in the stereotypiccondition (Hypothesis 4a), demonstrating more gender flexi-bility around play groups and less gender-based social exclu-sion in the counter-stereotypic condition. Finally, we expectedan interaction between participant gender and conditionwhereby in the stereotypic condition boys would report highergender-based social exclusion scores than girls would, due tostronger disapproval of cross-gender play. In the counter-stereotypic condition, we expected this difference to be atten-uated and expected boys and girls to show similar levels ofsocial exclusion (Hypothesis 4b). This is because the counter-stereotypic toy play of the children in this condition makes itless acceptable to exclude them and boys are more likely to beimpacted by this behavior.

Method

Participants

We recruited 96 British participants who were betweenthe ages of 4–8 years-old. Of this initial sample, 10 par-ticipants failed to complete all measures due to time con-straints and were not included in the final analysis. Inaddition, given that only four 8-year-olds completed thestudy, these data were also not included in the final anal-ysis due to minimal representation of this age group. Thefinal sample for analysis included 82 children (40 boysand 42 girls) aged 4–7 years-old (Mage = 5.4 years); girlsand boys did not significantly differ in age, t(80) = .21,p = .83. Participants were recruited from an urban primaryschool in a generally low SES neighborhood. The samplewas predominantly White, reflecting the low ethnic diver-sity in the area. Ethical consent was obtained from theResearch Ethics Committee at the University and we com-plied with British Psychological Society guidelines forresearch with children. Head teacher, parental, and partic-ipant consent were obtained prior to commencement ofthe study.

Procedure and Measures

Participants were told that they were going to be shown amagazine page which contained some pictures of childrenplaying with their favorite toys and that they would be askeda few questions about what they thought of the pictures.Participants were reminded that there were no right or wronganswers and that their answers were private. Once verbal con-sent had been obtained, participants were randomly assignedto the stereotypic or counter-stereotypic condition. In the ste-reotypic condition, participants viewed a magazine page fea-turing a male child playing with a car and female child playing

with a pony; those in the counter-stereotypic condition vieweda magazine page featuring a male child playing with a ponyand a female child playing with a car. Participants viewed themagazine page for 2 min. While the participant viewed themagazine page, the experimenter read aloud the followingwords from the page: BWe love it when you write to us withinteresting facts about your life, so this week we have askedour readers to send in photos of them playing with their favor-ite toys. Check out Sarah and Thomas’ photos below!^

Text in speech bubbles was presented next to the featuredmale and female children that the experimenter also readaloud. In the stereotypic condition with the female child, thespeech bubble read: BHello! My name is Sarah, and my favor-ite toy isMy Little Pony! I have lots, and play with them everyday.^ In the stereotypic condition with the male child, thespeech bubble read: BHello! My name is Thomas, and everyday I like to play with my cars. They’re my favorite toys!^ Inthe counter-stereotypic condition, the content of the speechbubbles was identical, but the children’s names were switchedso that BSarah^ liked to play with cars and BThomas^ liked toplay with My Little Pony. These pages are representative ofthose found in children’s magazines, where children’s photosand letters to the magazine are displayed, or the magazinepresents a feature on a reader.

Immediately after viewing the assigned magazine pages,participants completed a series of measures that assessed gen-der flexible attitudes and behavior. All study materials werepresented via Qualtrics on tablet computers. Participants com-pleted the measures individually with an experimenter in aquiet area.

Gender-Typed Toy Preferences

To assess gender flexible toy preferences, we presented par-ticipants with pictures of eight different toys, including fourstereotypically feminine toys (a wand, a pony, a baby doll, anda tea set) and four stereotypically masculine toys (a truck, a jetfighter plane, a tool kit, and a car), based on Blakemore andCenters’s (2005) categorization of toys as BStrongly FeminineToys^ and BStrongly Masculine Toys.^ The toys were pre-sented to participants individually and in a randomized order.We coded participants’ responses to the same question foreach of the eight toys: BHow much do you like this toy?^Participants selected from one of three response options basedon a scale depicting schematic faces: BNot at all^ (depictedwith a frowning face and coded as 1), BA little^ (depicted witha slightly smiling face and coded as 2), or BA lot^ (depictedwith a broadly smiling face and coded as 3). Total scores werecalculated separately for the feminine toys (α = .89) and mas-culine toys (α = .77) by summing the response for the fourtoys in each category separately. Scores for both types of toyscould range from 4 to 12, with higher scores indicating agreater preference for the respective toy type.

Sex Roles

Gender-Typed Toy Play

To assess gender flexible attitudes around toy play, we codedparticipants’ responses to the following question for each ofthe eight toys listed: BWho should play with this toy?^Participants selected from one of three response options,which were also paired with the corresponding gender sym-bols that appear on restroom signs: BOnly Girls^ (coded as 0),BOnly Boys^ (coded as 0), or BBoth Girls and Boys^ (codedas 1). Participants could respond verbally or by pointing to thesymbols of their choice (Weisgram et al. 2014). Total scoreswere calculated by summing the assigned codes across theeight toys. Scores could range from 0 to 8, with higher scoresindicating more gender flexible attitudes toward toy play. Itshould be noted that none of the participants indicated acounter-stereotypical endorsement (e.g., Bonly boys shouldplay with dolls^). This means that all responses coded as 0were stereotypical responses.

Gender-Typed Playmate Choice

To assess gender flexible attitudes in playmate choice, partic-ipants were presented with pictures of the children they hadviewed on the magazine pages (i.e. either the boy and girlengaged in stereotypic or counter-stereotypic toy play) andwere asked: BIf you had to choose one of the children to playwith, which one would you choose, the girl or the boy?^ Ifparticipants selected a girl playmate this was coded as 0; ifthey selected a boy playmate this was coded as 1. Afterselecting a playmate, we coded participants’ responses to thequestion: BWhy would you choose to play with this child?^Responses were coded into categories based on whether theyreferred to the gender label of the child pictured (coded 1), thetype of toy played with by the child pictured (coded 2), orsome other feature (coded as 3). It should be noted that noneof the participants referred to more than one category in theirresponses.

Gender-Based Social Exclusion

We adapted a measure from Killen and Stangor (2001) toassess gender flexible attitudes around social exclusion. Wepresented two scenarios to the participants, in a randomizedorder, and coded their responses. To assess the tendency toexclude the girl from boys’ play, we presented the followingscenario:

Imagine that a group of boys are playing with cars. Thisgirl [from the magazine page they viewed] comes overand asks if she can play. Two of the boys say that shecannot play because she is a girl. Is it alright or notalright for the boys to tell the girl that she can’t play?

To assess the tendency to exclude the boy from girls’ play, wepresented the following scenario:

Imagine that a group of girls are playing with dolls. Thisboy [from the magazine page they viewed] comes overand asks if he can play. Two of the girls say that hecannot play because he is a boy. Is it alright or not alrightfor the girls to tell the boy that he can’t play?

For each scenario, participants selected from one of three re-sponse options to indicate the extent to which they believed itwas all right to exclude the child from play: BNot alright^(coded 1), BA little bit alright^ (coded 2), or BAlright^ (coded3). A total gender-based exclusion score was computed bysumming the responses given for the two scenarios. Scoresranged from 2 to 6, with higher scores indicating thatgender-based social exclusion was more acceptable.

Results

Preliminary Analyses

Power analyses indicated that the statistical tests were suffi-ciently powered and the sample size was adequate for eachplanned analysis, with power to find an effect ranging be-tween 74% and 99% across all analyses (Howell 1992).Table 1 presents the overall means and standard deviationsfor the study variables, as well as the zero-order correlationsfor the associations among the continuous variables and point-biserial correlations for associations with the dichotomousvariable (i.e., gender-typed playmate choice). Correlationalanalyses were performed separately on the boys’ and girls’scores to examine initial relationships among the gender flex-ibility variables by gender group.

A significant positive association was observed betweenage and flexibility around gender-typed toy play for both boysand girls; as participants’ age increased, they were more likelyto believe that both boys and girls should play with both mas-culine and feminine toys (see Table 1). There was also a sig-nificant negative association between age and one’s owngender-typed toy preferences among boys and girls; as ageincreased, boys showed less interest in the masculine toysand girls showed less interest in the feminine toys. Analysesalso revealed a significant negative association between ageand acceptance of gender-based social exclusion, but onlyamong the boys; as age increased, boys showed less accep-tance of gender-based social exclusion across both conditions.No other variables correlated significantly with age. Giventhese associations with age, we included participants’ age asa covariate in our tests of the main gender flexibilityhypotheses.

Sex Roles

Several correlations were also observed among the genderflexibility variables for each gender group (see Table 1).Among boys, there was a significant positive relationship be-tween flexibility around gender-typed toy play and femininetoy preference scores, and a significant negative correlationbetween flexibility around gender-typed toy play and play-mate choice; as flexibility around toy play increased, so didthe likelihood that boys would choose a female playmate,across both conditions. There were no other significant corre-lations among the variables for boys. Among girls, analysesrevealed a significant negative relationship between flexibilityaround gender-typed toy play and interest in feminine toys,and a significant positive relationship between flexibilityaround gender-typed toy play and interest in masculine toys.There were no other significant correlations among the vari-ables for girls. Significant mean gender differences amongparticipants were observed only in the feminine and masculinetoy preference scores; these are reported in the following (alsosee Table 1).

Primary Analyses

Hypotheses for Gender-Typed Toy Preferences

We expected an interaction between participant gender andtoy type, whereby boys would prefer to play with mascu-line toys over feminine toys and girls would prefer to playwith feminine toys over masculine toys (Hypothesis 1a).We also expected a three-way interaction between partici-pant gender, condition, and toy type, whereby participantsin the counter-stereotypic condition would prefer other-gender toys more than children in the stereotypic condi-tion, demonstrating greater gender flexibility around toypreferences (Hypothesis 1b).

To test this set of hypotheses, we conducted a 2 (Condition:stereotypic vs. counter-stereotypic) × 2 (Participant Gender:girls vs. boys) × 2 (Toy Type: masculine vs. feminine) mixedanalysis of covariance (ANCOVA) on ratings of preferencefor masculine and feminine toys, with participant gender andcondition as the between-subjects factors, toy type as thewithin-subjects factor, and age entered as a covariate. In sup-port of Hypothesis 1a, we observed a significant interactionbetween participant gender and toy type, F(1, 75) = 197.55,MSE = 3.33, p < .001, ηp2 = .73. Pairwise comparisons re-vealed that girls preferred the feminine toys to the masculinetoys (p < .001, d = 2.21), and boys preferred the masculinetoys to the feminine toys (p < .001, d = 2.27; see Table 2).However, we did not observe support for Hypothesis 1b be-cause the three-way interaction among participant gender,condition, and toy type was not significant, F (1, 75) = 1.60,MSE = 3.33, p = .210, ηp2 = .02, suggesting that condition didnot affect children’s gender flexibility around toy preferences(see Table 2).

Hypotheses for Gender-Typed Toy Play

We expected a main effect of participant gender on gender-typed toy play, whereby girls would demonstrate more genderflexible attitudes toward toy play than boys would(Hypothesis 2a). We also expected a main effect for condition,whereby participants in the counter-stereotypic conditionwould be more likely to label toys as being for both boysand girls compared to participants in the stereotypic condition,demonstrating more gender flexible attitudes around toy play(Hypothesis 2b).

To test this set of hypotheses, we conducted a 2 (ParticipantGender) × 2 (Condition) between-subjects ANCOVA on atti-tudes toward gender-typed toy play, with age entered as acovariate. Refuting Hypothesis 2a, attitudes toward gender-

Table 1 Descriptive statistics and correlations for all study variables as a function of participants’ gender

Girls Boys Correlations

Variables M (SD) M (SD) 1 2 3 4 5 6

1. Age 5.38 (.99) 5.43 (.87) – −.55** .09 .56** .01 −.282. Gender-typed feminine toy preference 10.97 (1.39) 6.34 (2.55) −.25 – −.12 −.36* −.03 −.043. Gender-typed masculine toy preference 6.34 (2.55) 10.82 (1.39) −.36* −.12 – .35* −.04 .12

4. Flexibility in gender-typed toy play 3.51 (2.67) 2.74 (2.54) .41** .33* −.03 – .30 −.275. Gender-typed playmate choice .24 (.44) .70 (.46) −.12 −.27 −.16 −.42** – −.046. Gender-based social exclusion 2.57 (1.23) 3.03 (1.41) −.39* −.27 −.04 −.22 .13 –

Values for girls (n = 42) are presented above the diagonal; for boys (n = 40), below. Point-biserial correlations are reported for the associations with thedichotomous variable of gender-typed playmate choice, where 0 = girl playmate, and 1 = boy playmate. Higher scores indicate greater preference forgender-typed masculine and feminine toys and greater flexibility in gender-typed toy play, whereas higher scores for gender-based social exclusionindicate more exclusion of other-gender playmates, and therefore less flexibility in this domain. Higher scores for playmate choice indicate morepreference for a boy playmate

*p < .05. **p < .01

Sex Roles

typed toy play did not vary as a function of participants’ gen-der, F(1, 75) = 3.02, MSE = 5.03, p = .086, ηp2 = .04.However, in support of Hypothesis 2b, there was a significantmain effect of condition, F(1, 75) = 4.29, MSE = 5.03,p = .042, ηp2 = .05, whereby attitudes toward gender-typedtoy play were significantly more flexible among participantsin the counter-stereotypic condition (M = 3.64, SD = 2.70)compared to the stereotypic condition (M = 2.60, SD = 2.45).Participants, regardless of their own gender, were more likelyto endorse masculine toys and feminine toys as appropriate forboth boys and girls if they had viewed magazine contentdepicting children playing with counter-stereotypic toys.

Hypotheses for Gender-Typed Playmate Choice

We expected that participants would be more likely to choosea same-gender than an other-gender playmate in the stereotyp-ic condition, whereas we did not expect to observe this bias inthe counter-stereotypic condition (Hypothesis 3a), therebydemonstrating more gender flexible attitudes around playmatepreferences in the counter-stereotypic condition. Also in thecounter-stereotypic condition, we expected that boys wouldbe more likely than girls would be to choose an other-genderplaymate compared to a same-gender playmate (Hypothesis3b). We further expected that the reasons participants providefor their playmate choice would more likely refer to toy playstyle than to the playmate’s gender in the counter-stereotypiccondition, whereas we expected toy play style and playmate’sgender to be given as reasons in the stereotypic condition(Hypothesis 3c).

To test this set of hypotheses, we conducted two-way Chi-square tests with Yates correction for continuity to examinethe association between participant gender and gender-typedplaymate choice for each condition. In support of Hypothesis3a, in the stereotypic condition, girls were significantly morelikely to choose a same-gender playmate (91% vs. 9%) andboys were significantly more likely to choose a same-genderplaymate (94% vs. 6%) compared to an other-gender play-mate, χ2(1) = 26.51, p < .001, Cramer’s V = .85; however this

pattern was not observed in the counter-stereotypic condition,where girls (50% vs. 50%) and boys (45% vs. 55%) wereequally likely to select an other-gender versus same-genderplaymate, χ2(1) = .00, p = 1.00, Cramer’s V = .05. However,this finding refutes Hypothesis 3b, because in the counter-stereotypic condition, boys were not more likely than girls tochoose an other-gender over a same-gender playmate.

To examine participants’ reasoning behind their playmatepreferences, we conducted two one-sample Chi-square testsseparately for each condition. As we were primarily interestedin whether participants used the child’s play style or theirgender as a reason for choosing them as a playmate, reasonswhich did not fall into one of these two categories (classifiedas ‘other’) were excluded from analysis (16% of overall rea-sons in the counter-stereotypic condition; 31% in the stereo-typic condition). In support of Hypothesis 3c, participants inthe counter-stereotypic condition were significantly more like-ly to refer to the playmate’s play style (69%) than to the play-mate’s gender (31%) when choosing one of the playmates,χ2(1) = 4.50, p = .034, Cramer’s V = .38; however this patternwas not observed in the stereotypic condition, where partici-pants were not significantly more likely to refer to the play-mate’s play style (66%) over the playmate’s gender (33%),χ2(2) = 3.00, p = .083, Cramer’s V = .33.

Hypotheses for Gender-Based Social Exclusion

We expected a main effect for condition whereby participantsin the counter-stereotypic condition would report less endorse-ment of gender-based social exclusion than would participantsin the stereotypic condition (Hypothesis 4a). Finally, we ex-pected an interaction between participant gender and condi-tion whereby in the stereotypic condition boys would reporthigher gender-based social exclusion scores than girls would,due to stronger disapproval of cross-gender play. In thecounter-stereotypic condition we expected this difference tobe attenuated and expected boys and girls to show similarlevels of social exclusion (Hypothesis 4b).

Table 2 Gender-typed masculine and feminine toy preference scores as a function of condition and participant gender

(a) Three-way interaction (b) Two-way interaction

Stereotypic condition Counter-stereotypic condition Conditions combined

Participants’ gender n Feminine toysM (SD)

Masculine toysM (SD)

Feminine toysM (SD)

MasculinetoysM (SD)

Feminine toysM (SD)

Masculine toysM (SD)

Girls 42 11.05 (1.43) 7.68 (1.89) 10.89 (1.37) 6.95 (1.93) 10.97 (1.39)a 6.34 (1.92)bBoys 40 6.33 (2.77) 10.44 (1.54) 6.33 (2.42) 11.19 (1.17) 6.34 (2.55)a 10.82 (1.39)b

The mean differences (i.e., means with different subscripts in a row) between feminine and masculine toys for both female and male participants aresignificant (p < .001.)

Sex Roles

To test this set of hypotheses, we conducted a 2 (Condition)× 2 (Participant Gender) between-subjects ANCOVA ongender-based social exclusion scores, with age as a covariate.Counter to our expectations for Hypothesis 4a, the effect ofcondition was not significant, F(1, 77) = .25, MSE = 1.52,p = .620, ηp2 = .00. However, the analysis did reveal a signif-icant interaction between condition and participant gender,F(1, 77) = 4.59, MSE = 1.52, p = .035, ηp2 = .06. SupportingHypothesis 4b, pairwise comparisons revealed significantlyhigher endorsement of gender-based social exclusion amongboys (M = 3.27, SD = 1.63) compared to girls (M = 2.23,SD = .64) in the stereotypic condition (p = .008, d = .92), butnot between boys (M = 2.82, SD = 1.17) and girls (M = 2.96,SD = 1.66) in the counter-stereotypic condition (p = .733,d = .06), suggesting some attenuation of boys’ gender-typedbiases around play style and playmates if they are alreadyaware that the child possesses a counter-stereotypic play style.However, pairwise comparisons examining differences ingirls’ and boys’ gender-based social exclusion scores acrossconditions were not significant (ps > .05).

Discussion

The present study investigated the impact of stereotypic andcounter-stereotypic children presented in children’s maga-zines on participants’ gender flexibility around gender-typedtoy preferences for themselves and others, playmate choices,and endorsement of gender-based social exclusion. Most ofthe hypotheses were fully supported and a number of impor-tant patterns were observed. Participants did not demonstratemore gender flexible attitudes in their own preferences forgender-typed toys (i.e., girls preferred feminine toys and boyspreferred masculine toys) after exposure to counter-stereotypic content. However, participants in the counter-stereotypic condition did demonstrate more gender flexibleattitudes toward the toy play of other boys and girls, labelingmasculine toys and feminine toys as appropriate for both boysand girls more often than participants in the stereotypiccondition.

In addition, we observed a strong preference for same-gender playmates over other-gender playmates among partic-ipants in the stereotypic condition, but we observed no pref-erence for same-gender playmates over other-gender play-mates among participants in the counter-stereotypic condition.This choice of playmate in the counter-stereotypic conditionappeared to be driven more by the type of toy play beingmodeled by the child than by the child’s gender. Using a moreexplicit indicator of social exclusion, we found that in thestereotypic condition, boys were more supportive of gender-based exclusion than were girls. Meanwhile in the counter-stereotypic condition, there were no significant differences

between boys and girls in their endorsement of gender-basedexclusion.

Our study represents the first known investigation of theimpact of counter-stereotypic peers pictured in children’smagazines on children’s gender flexibility around toy andplaymate attitudes and preferences. On the whole, our findingssuggest that exposure to counter-stereotypic content that chal-lenges gender-typed toy play may be a useful strategy forattenuating gender-typed attitudes and behavior in young chil-dren, at least encouraging more flexible thinking around thegender-typed toy play of other boys and girls.

In contrast to Green et al. (2004) who used fictional char-acters to display counter-stereotypic gender models, we foundthat gender counter-stereotypic peers shifted boys’ as well asgirls’ gender-typed attitudes. This suggests that pictured ex-amples of actual children engaged in counter-stereotypic toyplay (in a media format) may be more effective at changingchildren’s gender-typed attitudes than the use of fictional char-acters. The use of actual children may also facilitate greaterperceived behavioral similarity with the peers, which has beenlinked to the potential countering of gender-typed attitudesand behaviors (Martin et al. 2011). Children may have per-ceived themselves as similar to the other-gender peer in thecounter-stereotypic condition if the peer displayed similar toypreferences to themselves, and this possibility should be ex-plored further in future research.

Although we hypothesized that exposure to the counter-stereotypic peers in the magazine would amplify children’sgender flexibility, it is perhaps unsurprising that children’sown preferences for toys remained gender-typed. This patternis consistent with previous research, which has shown chil-dren’s gender attitudes are easier to manipulate than their be-haviors (for example, Bigler and Liben 1990, in the context ofgender-typed occupations). Children’s own gender-related at-titudes may be less flexible because of the increased risk ofpeer rejection associated with preferences (and behaviors) thatbreak gender norms. Therefore, more intensive interventionswith peer reinforcement may be required to effectively changechildren’s own gender-typed toy preferences. This under-scores the idea that a single exposure to gender atypical toyplay would not affect deeply entrenched attitudes (Weisgramet al. 2014; Wong and Hines 2015), a point to which we returnin the Limitations section.

The fact that such exposure did alter attitudes aroundother’s toy play was consistent with expectations and warrantsfurther consideration. There is some evidence to suggest thatexposure to non-traditional toy play in television commercialscan increase gender flexible attitudes around toy play in chil-dren between 6 and 8 years-old (Pike and Jennings 2005).Given the role of toy play style in directing children’s socialand cognitive development (Alexander 2003; Alexander andHines 2002; Blakemore et al. 2009; Blakemore and Centers2005; De Lisi and Wolford 2002), it behooves scholars and

Sex Roles

practitioners to understand how we can harness toy play tomaximize potential and growth for all children. Furthermore,engagement with a wider variety of toys that cross traditionalgender lines may increase the possibility for more cross-gender friendships to develop and be sustained, which has alsoshown to be beneficial for children’s development (Fabes et al.2003).

Limitations and Future Research Directions

Although we reported some interesting results regarding thegender flexibility of young children, our study is not withoutits limitations. First, we recognize the impact of the presentstudy may be limited by the fact that we did not include acontrol condition against which to compare the direction ofthe observed effects. In future research, we would recommenda comparison against a peer playing with a gender-neutral toy(e.g., a puzzle) as well as against a non-exposure condition,which would reflect a truer baseline for gender flexible atti-tudes and behaviors. Furthermore, future research should stan-dardize the images of the peers across conditions. Future re-search should also directly compare exposure to life-like peerswith storybook characters to examine whether these imagesaffect gender flexibility to different extents.

Second, we examined behavioral intentions in the contextof hypothetical scenarios and contrived stimuli presented tochildren. An important next step in this program of researchwould be to examine toy and playmate preferences in thecontext of viewing gender counter-stereotypic peers in maga-zines in more natural settings where actual behavior can beobserved.

Third, we presented participants with one exposure to asingle magazine page and the impact on gender attitudes andpreferences was measured immediately afterwards. This de-sign was adopted with the intention of providing a snapshot ofhow media can impact children’s gender attitudes and prefer-ences. Clearly a more intensive and regular intervention usingcounter-stereotypic peers would be necessary for long-lastingchange, and future research is required in order to examine thelong-term impact of such an intervention program. Such re-search would also determine whether the effects observed aredue to priming or to more substantial changes to children’sunderstanding of and adherence to gender stereotypes.

Fourth, our relatively small sample size limited our abilityto detect small andmoderate effects of the magazine exposure,and it also precluded us from making age-based comparisons.Analyses revealed that gender flexibility as measured bygender-typed toy play, own gender-typed toy preference, andgender-based exclusion (boys only) was correlated with age.However, between the ages of 4–7 years, children undergoconsiderable changes in their understanding of and adherenceto gender stereotypes. Rigidity and adherence to stereotypesappears to increase linearly from 3 to 6 years-old and begins to

decline thereafter when gender flexibility emerges (Serbin andSprafkin 1986; Signorella et al. 1993; Trautner et al. 2005).Future research should include a larger sample size to allow athorough examination of developmental changes in behavior-al (e.g. social exclusion, playmate preference) and attitudinalaspects of gender flexibility, as well as how these change inresponse to the peers across the age group studied here (i.e.,compare the impact of the peers among children aged 4-, 5-, 6-and 7-years-old). Future research could also include measuresof understanding of gender, such as gender constancy, to cap-ture the differential impact of the peers depending on thechild’s stage of gender development.

Moreover, future research should also examine the differ-ential effect of exposure to gender stereotypic and counter-stereotypic children pictured in media across this age range.For example, research on encoding and memory processes hasfound that young childrenmisremember or incorrectly processgender counter-stereotypic information to match their pre-existing gender schema (Liben and Signorella 1980; Martinand Halverson 1983; Signorella and Liben 1984). This re-search would suggest that a single exposure to gendercounter-stereotypic children pictured in magazines or othermedia would have a stronger impact among the older childrenin our sample. This possibility warrants further study.

We also limited the playmate choice and social exclusionmeasures to ask about the children presented to the partici-pants in the magazine. This was done in order to enhance therealistic nature of the scenarios, but it did mean the specificchildren targeted in the social exclusion scenarios varied bycondition. Future research could adapt the methods employedhere to include a variety of social exclusion scenarios, withnew targets in the social exclusion scenario, in addition tothose viewed in the magazines, in order to improve experi-mental control and test the generalizability of this finding tonew children and social situations. It would also be beneficialto include an additional response option of Bboth^ in the play-mate choice measure to allow children to express a preferencefor playing with both girls and boys, instead of restricting theirresponse to choosing one gender over the other, which may bemasking children’s gender flexible preferences. The PPPSI(Pasterski et al. 2011) could also be included in future researchto gain more detailed information about children’s play styleand playmate preferences beyond what the present study wasable to obtain.

It is noteworthy that the effect of counter-stereotypic peerson participant’s attitudes toward gender-typed toy play andplaymate preference was the same across boys and girls. Infact, for endorsement of gender-based exclusion, counter-stereotypic peers brought boys’ and girls’ attitudes more inline with one another. This attenuation of gender bias in youn-ger boys is, therefore, especially revealing. It could be arguedthat it is easier to change girls’ gender-related attitudes andbehaviors, compared to boys’, because greater resistance to

Sex Roles

gender-atypical attitudes and behavior is more often observedamong boys (Bussey and Perry 1982; Mulvey and Killen2015). Furthermore, masculine traits, toys, and behaviors aregenerally accepted as being of higher status than their femi-nine counterparts, and research has shown that children areaware of these status differences (Ruble et al. 2006. For thatreason, it may be easier to persuade girls to move towardmasculine toy preferences, for instance, than persuading boysto choose feminine toys (Blakemore 2003, Horn 2008).Because boys and girls responded to the counter-stereotypicpeers similarly in the present study, the use of pictures ofactual children in a magazine format may overcome some ofthe difficulties in adjusting boys’ gender-typed attitudes.However, it would be interesting for future studies to examinethe social status associated with masculine and feminine toysand activities in tempering the impact of stereotypic andcounter-stereotypic peers on children’s gender flexibility.

Practice Implications

The findings of the present study have several implications.First, these findings suggest it is possible to shift childrentoward more gender flexible attitudes and change children’sviews on gender-related play. This possibility counters laybeliefs that gender segregation and gendered toy preferenceis inevitable in young children, and it adds to literature em-phasizing the potential for change in children’s attitudes aboutgender-related play and friendships.

In particular, our findings suggest more regular exposure tocounter-stereotypic content in the media could be an effectivestrategy to promote gender flexibility and combat gender-related bullying (Bigler 1999; Bussey and Perry 1982). Thatsuch an acute exposure shifted attitudes, underscores the im-pact that repeated exposure to gender stereotypical media canhave on young children. These findings suggest that present-ing children with images of counter-stereotypic peers throughmagazines could be used to encourage children to play withtheir own and other-gender toys, play inmixed-gender groups,and reduce gender-based social exclusion and bullying forboth gender-typical and gender-atypical children.

Educators, parents, and policymakers might benefit fromthe present research and the approach tested to increase genderflexibility in children. This exposure technique could be ex-tended for use in the classroom by providing more regularexposure to counter-stereotypic peers in children’s mediathrough a series of magazine articles, or news stories, thatfeature such children. Children could also be asked to modeland create their own resources. Moreover, our research showsthat children consider both play style and gender whenselecting a playmate. This finding suggests that highlightingbehavioral similarities in children could encourage mixed-gender play. We suggest encouraging mixed-gender play byteachers and parents, despite the apparent gender segregation

during play, because boys and girls are willing to play withone another if they possess similar toy and play style interests.

Conclusion

Exposure to gender counter-stereotypic peers in a magazineformat increases gender flexibility among young children.Specifically, children exposed to counter-stereotypic peerswere more flexible in their attitudes toward what other chil-dren could play with and were more likely to choose an other-gender playmate, using play style as a guide more so than theplaymate’s gender. Moreover, boys’ stronger endorsement ofsocial exclusion in the stereotypic condition was attenuated inthe counter-stereotypic condition. The results of the presentstudy not only underscore the impact of media (specificallyprint media) on children’s early understanding of gender andconformity to gender stereotypes, but also highlight the poten-tial use of media to challenge and disrupt gender-typed toychoices and playgroups in young children. In particular, thisresearch highlights the potential use of counter-stereotypicsame-age peers in children’s print media to normalizecounter-stereotypic attitudes, and perhaps behaviors, as an im-portant avenue for future research and intervention. On thewhole, these results suggest that the observed play style andtoy preferences of others could be used as a gateway to genderdesegregation in children. We hope the present study willinspire further investigations of this possibility in children.

Compliance with Ethical Standards We, the authors, confirm that ethicalconsent was obtained from the Research Ethics Committee at theUniversity of Kent and we complied with British Psychological Societyguidelines for research with children. Head teacher, parental, and partic-ipant consent were obtained prior to commencement of the study, andethical guidelines were adhered to throughout.

Open Access This article is distributed under the terms of the CreativeCommons At t r ibut ion 4 .0 In te rna t ional License (h t tp : / /creativecommons.org/licenses/by/4.0/), which permits unrestricted use,distribution, and reproduction in any medium, provided you give appro-priate credit to the original author(s) and the source, provide a link to theCreative Commons license, and indicate if changes were made.

References

Abad, C., & Pruden, S. M. (2013). Do storybooks really break children'sgender stereotypes? Frontiers in Psychology, 4(986), 1–4. https://doi.org/10.3389/fpsyg.2013.00986.

Alexander, G. M. (2003). An evolutionary perspective of sex-typed toypreferences: Pink, blue, and the brain. Archives of Sexual Behavior,32(1), 7–14. https://doi.org/10.1023/A:1021833110722.

Alexander, G. M., & Hines, M. (2002). Sex differences in response tochildren's toys in nonhuman primates. Evolution and HumanBehavior, 23(6), 467–479. https://doi.org/10.1016/S1090-5138(02)00107-1.

Sex Roles

Ashton, E. (1983). Measures of play behavior: The influence of sex-rolestereotyped children’s books. Sex Roles, 9(1), 43–47. https://doi.org/10.1007/BF00303108.

Aubry, S., Ruble, D. N., & Silverman, L. B. (1999). The role of genderknowledge in children's gender-typed preferences. In L. Balter, C. S.Tamis-LeMonda, L. Balter, & C. S. Tamis-LeMonda (Eds.), Childpsychology: A handbook of contemporary issues (pp. 363–390).New York: Psychology Press.

Auster, C. J., & Mansbach, C. S. (2012). The gender marketing of toys:An analysis of color and type of toy on the Disney store website. SexRoles, 67(7–8), 375–388. https://doi.org/10.1007/s11199-012-0177-8.

Bakir, A. (2013). Character portrayal: Examining gender roles in televi-sion commercials targeted at children in India and the United States.Journal of Global Marketing, 26(2), 57–67. https://doi.org/10.1080/08911762.2013.804612.

Bakir, A., & Palan, K. M. (2013). Agentic and communal: Multinationalanalysis of gender portrayal in children's television commercials.Journal of Current Issues and Research in Advertising, 34(1), 39–56. https://doi.org/10.1080/10641734.2013.754708.

Bartini, M. (2006). Gender role flexibility in early adolescence:Developmental change in attitudes, self-perceptions, and behaviors.Sex Roles, 55(3–4), 233–245. https://doi.org/10.1007/s11199-006-9076-1.

Bauer, P. J. (1993). Memory for gender-consistent and gender-inconsistent event sequences by twenty-five-month-old children.Child Development, 64(1), 285–297. https://doi.org/10.2307/1131452.

Bem, S. (1981). Gender schema theory: A cognitive account of sex typ-ing. Psychological Review, 88, 325–364.

Bem, S. L. (1983). Gender schema theory and its implications for childdevelopment: Raising gender-aschematic children in a gender-schematic society. Signs, 8(4), 598–616.

Bigler, R. S. (1999). Psychological interventions designed to countersexism in children: Empirical limitations and theoretical founda-tions. In W. B. Swann Jr. & J. H. Langlois (Eds.), Sexism and ste-reotypes in modern society: The gender science of Janet TaylorSpence (pp. 129–151). Washington, DC: American PsychologicalAssociation. https://doi.org/10.1037/10277-006.

Bigler, R. S., & Liben, L. S. (1990). The role of attitudes and interventionsin gender-schematic processing. Child Development, 61(5), 1440–1452. https://doi.org/10.2307/1130754.

Blakemore, J. O. (2003). Children's beliefs about violating gender norms:Boys shouldn't look like girls, and girls shouldn't act like boys. SexRoles, 48(9–10), 411–419. https:/ /doi.org/10.1023/A:1023574427720.

Blakemore, J. O., & Centers, R. E. (2005). Characteristics of boys’ andgirls’ toys. Sex Roles, 53(9–10), 619–633. https://doi.org/10.1007/s11199-005-7729-0.

Blakemore, J. O., Berenbaum, S. A., & Liben, L. S. (2009). Genderdevelopment. New York: Psychology Press.

Boyd, H., & Murnen, S. K. (2017). Thin and sexy vs. muscular anddominant: Prevalence of gendered body ideals in popular dolls andaction figures. Body Image, 21, 90–96. https://doi.org/10.1016/j.bodyim.2017.03.003.

Bradbard, M. R., & Endsley, R. C. (1983). The effects of sex-typedlabeling on preschool children's information-seeking and retention.Sex Roles, 9(2), 247–260.

Buhs, E. S., Ladd, G. W., & Herald, S. L. (2006). Peer exclusion andvictimization: Processes that mediate the relation between peergroup rejection and children's classroom engagement and achieve-ment? Journal of Educational Psychology, 98(1), 1–13. https://doi.org/10.1037/0022-0663.98.1.1.

Bussey, K., & Bandura, A. (1999). Social cognitive theory of genderdevelopment and differentiation. Psychological Review, 106(4),676–713. https://doi.org/10.1037/0033-295X.106.4.676.

Bussey, K., & Bandura, A. (2004). Social cognitive theory of genderdevelopment and functioning. In A. H. Eagly, A. E. Beall, & R. J.Sternberg (Eds.), The psychology of gender (pp. 92–119). NewYork: Guilford Press.

Bussey, K., & Perry, D. G. (1982). Same-sex imitation: The avoidance ofcross-sex models or the acceptance of same-sex models? Sex Roles,8(7), 773–784.

Cahill, B., & Adams, E. (1997). An exploratory study of early childhoodteachers’ attitudes toward gender roles. Sex Roles, 36(7–8), 517–529. https://doi.org/10.1007/BF02766688.

Caldera, Y., Huston, A., & O'Brien, M. (1989). Social interactions andplay patterns of parents and toddlers with feminine, masculine, andneutral toys. Child Development, 60(1), 70–76. https://doi.org/10.2307/1131072.

Campbell, A., Shirley, L., & Caygill, L. (2002). Sex-typed preferences inthree domains: Do two-year-olds need cognitive variables? BritishJournal of Psychology, 93(2), 203–217. https://doi.org/10.1348/000712602162544.

Cherney, I. D., & London, K. (2006). Gender-linked differences in thetoys, television shows, computer games, and outdoor activities of 5-to 13-year-old children. Sex Roles, 54(9–10), 717–726. https://doi.org/10.1007/s11199-006-9037-8.

Cimpian, A., Mu, Y., & Erickson, L. C. (2012). Who is good at thisgame? Linking an activity to a social category undermines children’sachievement. Psychological Science, 23(5), 533–541. https://doi.org/10.1177/0956797611429803.

Collins, R. L. (2011). Content analysis of gender roles in media: Whereare we now and where should we go? Sex Roles, 64(3–4), 290–298.https://doi.org/10.1007/s11199-010-9929-5.

Coyne, S. M., Linder, J. R., Rasmussen, E. E., Nelson, D. A., & Collier,K. M. (2014). It’s a bird! It’s a plane! It’s a gender stereotype!:Longitudinal associations between superhero viewing and genderstereotyped play. Sex Roles, 70(9–10), 416–430. https://doi.org/10.1007/s11199-014-0374-8.

Coyne, S.M., Linder, J. R., Rasmussen, E. E., Nelson, D. A., & Birkbeck,V. (2016). Pretty as a princess: Longitudinal effects of engagementwith Disney princesses on gender stereotypes, body esteem, andprosocial behavior in children. Child Development, 87(6), 1909–1925. https://doi.org/10.1111/cdev.12569.

De Lisi, R., &Wolford, J. L. (2002). Improving children's mental rotationaccuracy with computer game playing. The Journal of GeneticPsychology, 163(3), 272–282. https://doi.org/10.1080/00221320209598683.

Dinella, L. M. (2017). Halloween costume choices: Reflections of genderdevelopment in early childhood. The Journal of GeneticPsychology: Research and Theory on Human Development,178(3), 165–178. https://doi.org/10.1080/00221325.2017.1295223.

Dittmar, H., Halliwell, E., & Ive, S. (2006). Does Barbie make girls wantto be thin? The effect of experimental exposure to images of dolls onthe body image of 5-to 8-year-old girls. Developmental Psychology,42(2), 283–292. https://doi.org/10.1037/0012-1649.42.2.283.

Fabes, R. A.,Martin, C. L., &Hanish, L. D. (2003). Young children's playqualities in same-, other-, and mixed-sex peer groups. ChildDevelopment, 74(3), 921–932. https://doi.org/10.1111/1467-8624.00576.

Foster, M. C. (2016). Representations of gender and embodiment inchildren's picture books. (Doctoral thesis, University of Toronto,Canada). Retrieved from PsycINFO database. (2016-17338-234).

Geary, D. C., & Bjorklund, D. F. (2000). Evolutionary developmentalpsychology. Child Development, 71(1), 57–65. https://doi.org/10.1111/1467-8624.00118.

Gerbner, G. (1998). Cultivation analysis: An overview. MassCommunication and Society, 1(3–4), 175–194.

Gould, M. (2008, October 3). Girls choosing camera lenses over micro-scopes. The Guardian. Retrieved from https://www.theguardian.com/education/2008/oct/03/science.choosingadegree.

Sex Roles

Government Equalities Office, UK. (2016). UK gender pay gap.Retrieved from https://www.gov.uk/government/news/uk-gender-pay-gap.

Green, V. A., Bigler, R., & Catherwood, D. (2004). The variability andflexibility of gender-typed toy play: A close look at children's be-havioral responses to counter-stereotypic models. Sex Roles, 51(7–8), 371–386. https://doi.org/10.1023/B:SERS.0000049227.05170.aa.

Halim, M. L., Ruble, D., Tamis-LeMonda, C., & Shrout, P. E. (2013).Rigidity in gender-typed behaviors in early childhood: A longitudi-nal study of ethnic minority children. Child Development, 84(4),1269–1284. https://doi.org/10.1111/cdev.12057.

Halim,M. L., Ruble, D. N., Tamis-LeMonda, C. S., Zosuls, K.M., Lurye,L. E., &Greulich, F. K. (2014). Pink frilly dresses and the avoidanceof all things ‘girly’: Children’s appearance rigidity and cognitivetheories of gender development. Developmental Psychology,50(4), 1091–1101. https://doi.org/10.1037/a0034906.

Hofstede, G. H. (2001). Culture’s consequences: Comparing values, be-haviors, institutions, and organizations across nations. ThousandOaks: Sage.

Horn, S. S. (2008). The multifaceted nature of sexual prejudice: Howadolescents reason about sexual orientation and sexual prejudice.In S. R. Levy, M. Killen, S. R. Levy, & M. Killen (Eds.),Intergroup attitudes and relations in childhood through adulthood(pp. 173–188). New York: Oxford University Press.

Howell, D. C. (1992). Statistical methods for psychology (3rd ed.).Boston: Duxbury Press.

Hunt, B. (2004, October 28). Surge in number of female-led enterprises.Financial Times (UK edition). Retrieved from http://www.ft.com/cms/s/0/64efe4f4-287e-11d9-9308-00000e2511c8.html?ft_site=falcon&desktop=true#axzz4fZABtGoz.

Huston, A. C. (1983). Sex-typing. In E. M. Hetherington (Ed.),Handbook of child psychology: Socialization, personality, and so-cial development (Vol. 4, pp. 388–467). New York: Wiley.

Jirout, J. J., & Newcombe, N. S. (2015). Building blocks for developingspatial skills: Evidence from a large, representative U.S. sample.Psychological Science, 26(3), 302–310. https://doi.org/10.1177/0956797614563338.

Kahlenberg, S. G., & Hein, M. M. (2010). Progression on nickelodeon?Gender-role stereotypes in toy commercials. Sex Roles, 62(11–12),830–847. https://doi.org/10.1007/s11199-009-9653-1.

Karniol, R., & Gal-Disegni, M. (2009). The impact of gender-fair versusgender-stereotyped basal readers on 1st-grade children's gender ste-reotypes: A natural experiment. Journal of Research in ChildhoodEducat ion, 23 (4) , 411–420. ht tps : / /doi .org/10.1080/02568540909594670.

Katz, P. A., & Ksansnak, K. R. (1994). Developmental aspects of genderrole flexibility and traditionality in middle childhood and adoles-cence. Developmental Psychology, 30(2), 272. https://doi.org/10.1037/0012-1649.30.2.272.

Killen, M., & Stangor, C. (2001). Children's social reasoning about in-clusion and exclusion in gender and race peer group contexts. ChildDevelopment, 72(1), 174–186. https://doi.org/10.1111/1467-8624.00272.

Kornienko, O., Santos, C. E.,Martin, C. L., &Granger, K. L. (2016). Peerinfluence on gender identity development in adolescence.Developmental Psychology, 52(10), 1578–1592. https://doi.org/10.1037/dev0000200.

Leaper, C. (1994). Exploring the consequences of gender segregation onsocial relationships. In C. Leaper & C. Leaper (Eds.), Childhoodgender segregation: Causes and consequences (pp. 67–86). SanFrancisco: Jossey-Bass.

Leaper, C., Breed, L., Hoffman, L., & Perlman, C. A. (2002). Variationsin the gender-stereotyped content of children's television cartoonsacross genres. Journal of Applied Social Psychology, 32(8), 1653–1662. https://doi.org/10.1111/j.1559-1816.2002.tb02767.x.

Li, R. H., &Wong,W. I. (2016). Gender-typed play and social abilities inboys and girls: Are they related? Sex Roles, 74(9–10), 399–410.https://doi.org/10.1007/s11199-016-0580-7.

Liben, L. S., & Signorella, M. L. (1980). Gender-related schemata andconstructive memory in children. Child Development, 51(1), 11–18.https://doi.org/10.2307/1129584.

Liben, L. S., Bigler, R. S., & Krogh, H. R. (2002). Language at work:Children's gendered interpretations of occupational titles. ChildDevelopment, 73(3), 810–828. https://doi.org/10.1111/1467-8624.00440.

LoBue, V., & DeLoache, J. S. (2011). Pretty in pink: The early develop-ment of gender-stereotyped colour preferences. British Journal ofDevelopmental Psychology, 29(3), 656–667. https://doi.org/10.1111/j.2044-835X.2011.02027.x.

Maccoby, E. E. (2002). Gender and group process: A developmentalperspective. Current Directions in Psychological Science, 11(2),54–58. https://doi.org/10.1111/1467-8721.00167.

Maccoby, E. E., & Jacklin, C. N. (1987). Gender segregation in child-hood. Advances in Child Development and Behavior, 20, 239–287.https://doi.org/10.1016/S0065-2407(08)60404-8.

Martin, C. L., & Fabes, R. A. (2001). The stability and consequences ofyoung children’s same-sex peer interactions. DevelopmentalPsychology, 37(3), 431–446. https://doi.org/10.1037/0012-1649.37.3.431.

Martin, C. L., & Halverson, C. F. (1983). The effects of sex-typingschemas on young children’s memory. Child Development, 54(3),563–574. https://doi.org/10.2307/1130043.

Martin, C. L., & Ruble, D. (2004). Children's search for gender cues:Cognitive perspectives on gender development. Current Directionsin Psychological Science, 13(2), 67–70. https://doi.org/10.1111/j.0963-7214.2004.00276.x.

Martin, C. L., Fabes, R. A., Hanish, L., Leonard, S., & Dinella, L. M.(2011). Experienced and expected similarity to same-gender peers:Moving toward a comprehensive model of gender segregation. SexRoles, 65(5–6), 421–434. https://doi.org/10.1007/s11199-011-0029-y.

Merskin, D. (2002). Boys will be boys: A content analysis of gender andrace in children's advertisements on the Turner Cartoon Network.Journal of Current Issues and Research in Advertising, 24(1), 51–59. https://doi.org/10.1080/10641734.2002.10505127.

Miller, M. K., & Summers, A. (2007). Gender differences in video gamecharacters’ roles, appearances, and attire as portrayed in video gamemagazines. Sex Roles, 57(9–10), 733–742. https://doi.org/10.1007/s11199-007-9307-0.

Miller, C. F., Trautner, H. M., & Ruble, D. N. (2006). The role of genderstereotypes in children’s preferences and behavior. In L. Balter & C.S. Tamis-LeMonda (Eds.), Child psychology: A handbook of con-temporary issues (pp. 293–323). New York: Psychology Press.

Mulvey, K. L., & Killen, M. (2015). Challenging gender stereotypes:Resistance and exclusion. Child Development, 86(3), 681–694.https://doi.org/10.1111/cdev.12317.

Murnen, S. K., Greenfield, C., Younger, A., & Boyd, H. (2016). Boys actand girls appear: A content analysis of gender stereotypes associatedwith characters in children’s popular culture. Sex Roles, 74(1–2),78–91. https://doi.org/10.1007/s11199-015-0558-x.

Nelson, A. (2005). Children’s toy collections in Sweden—A less gender-typed country? Sex Roles, 52(1–2), 93–102. https://doi.org/10.1007/s11199-005-1196-5.

Nhundu, T. J. (2007). Mitigating gender-typed occupational preferencesof Zimbabwean primary school children: The use of biographicalsketches and portrayals of female role models. Sex Roles, 56(9–10),639–649. https://doi.org/10.1007/s11199-007-9204-6.

Pasterski, V., Geffner, M. E., Brain, C., Hindmarsh, P., Brook, C., &Hines, M. (2011). Prenatal hormones and childhood sex segrega-tion: Playmate and play style preferences in girls with congenital

Sex Roles

adrenal hyperplasia. Hormones and Behavior, 59(4), 549–555.https://doi.org/10.1016/j.yhbeh.2011.02.007.

Perry, D. G., & Bussey, K. (1979). The social learning theory of sexdifferences: Imitation is alive and well. Journal of Personality andSocial Psychology, 37(10), 1699–1712. https://doi.org/10.1037/0022-3514.37.10.1699.

Pike, J. J., & Jennings, N. A. (2005). The effects of commercials onchildren's perceptions of gender appropriate toy use. Sex Roles,52(1–2), 83–91. https://doi.org/10.1007/s11199-005-1195-6.

Pingree, S. (1978). The effects of nonsexist television commercials andperceptions of reality on children's attitudes about women.Psychology of Women Quarterly, 2(3), 262–277.

Plakoyiannaki, E., & Zotos, Y. (2009). Female role stereotypes in printadvertising: Identifying associations with magazine and product cat-egories. European Journal of Marketing, 43(11–12), 1411–1434.https://doi.org/10.1108/03090560910989966.

Rich, E. (2005). Young women, feminist identities and neoliberalism.Women’s Studies International Forum, 28, 495–508. https://doi.org/10.1016/j.wsif.2005.09.006.

Ruble, D. N., &Martin, C. L. (1998). Gender development. InW. Damon& N. Eisenberg (Eds.), Handbook of child psychology: Social, emo-tional, and personality development (Vol. 3, pp. 933–1016). NewYork: Wiley.

Ruble, D. N., Martin, C. L., & Berenbaum, S. (2006). Gender develop-ment. In W. Damon & N. Eisenberg (Eds.), Handbook of childpsychology: Personality and social development (Vol. 3, 6th ed.,pp. 858–932). New York: Wiley.

Schuette, C. T., Ponton, M. K., & Charlton, M. L. (2012). Middle schoolchildren's career aspirations: Relationship to adult occupations andgender. The Career Development Quarterly, 60(1), 35–46. https://doi.org/10.1002/j.2161-0045.2012.00004.x.

Scott, K. P., & Feldman-Summers, S. (1979). Children's reactions totextbook stories in which females are portrayed in traditionally maleroles. Journal of Educational Psychology, 71(3), 396–402. https://doi.org/10.1037/0022-0663.71.3.396.

Serbin, L. A., & Sprafkin, C. (1986). The salience of gender and theprocess of sex typing in three- to seven-years old children. ChildDevelopment, 57(5), 1188–1199. https://doi.org/10.2307/1130442.

Serbin, L. A., Moller, L. C., Gulko, J., Powlishta, K. K., & Colboume, K.A. (1994). The emergence of gender segregation in toddlerplaygroups. In C. Leaper (Ed.), Childhood gender segregation:Causes and consequences (pp. 7–18). San Francisco: Jossey- Bass.

Sheldon, J. P. (2004). Gender stereotypes in educational software foryoung children. Sex Roles, 51(7–8), 433–444. https://doi.org/10.1023/B:SERS.0000049232.90715.d9.

Shell, R., & Eisenberg, N. (1990). The role of peers’ gender in children'snaturally occurring interest in toys. International Journal ofBehavioral Development, 13(3), 373–388. https://doi.org/10.1177/016502549001300309.

Sherman, A. M., & Zurbriggen, E. L. (2014). BBoys can be anything^:Effect of Barbie play on girls’ career cognitions. Sex Roles, 70(5–6),195–208. https://doi.org/10.1007/s11199-014-0347-y.

Shutts, K., Banaji, M. R., & Spelke, E. S. (2010). Social categories guideyoung children’s preferences for novel objects. DevelopmentalScience, 13(4), 599–610. https://doi.org/10.1111/j.1467-7687.2009.00913.x.

Signorella, M. L., & Liben, L. S. (1984). Recall and reconstruction ofgender-related pictures: Effects of attitude, task difficulty, and age.Child Development, 55(2), 393–405. https://doi.org/10.2307/1129951.

Signorella, M. L., Bigler, R. S., & Liben, L. S. (1993). Developmentaldifferences in children’s gender schemata about others: A meta-analytic review. Developmental Review, 13(2), 147–183. https://doi.org/10.1006/drev.1993.1007.

Signorielli, N. (2001). Television’s gender role images and contribution tostereotyping: Past, present, future. In D. Singer & J. Singer (Eds.),Handbook of children and the media (pp. 341–358). ThousandOaks: Sage.

Skinner, C. (2013). Gender and language in best-selling children's pic-ture books: Who gets to speak? (Doctoral dissertation). Retrievedfrom PsychINFO database. (2013-99240-205).

Slaby, R. G., & Frey, K. S. (1975). Development of gender constancy andselective attention to same-sex models. Child Development, 46(4),849–856. https://doi.org/10.2307/1128389.

Spinner, L., Cameron, L., & Tenenbaum, H. (2016). Gender-stereotypedmessages in pre-school children's magazines. Presentation,University of Kent.

Statista. (2016a). Retrieved from https://www.statista.com/statistics/321625/children-s-magazines-ranked-by-sales-volume-uk-2013/.

Statista. (2016b). Retrieved from https://www.statista.com/statistics/398258/children-use-of-books-magazines-and-comics-by-age-uk/.

Steyer, I. (2014). Gender representations in children's media and theirinfluence. Campus-Wide Information Systems, 31(2/3), 171–280.https://doi.org/10.1108/CWIS-11-2013-0065.

Thompson, T. L., & Zerbinos, E. (1995). Gender roles in animated car-toons: Has the picture changed in 20 years? Sex Roles, 32(9–10),651–673. https://doi.org/10.1007/BF01544217.

Trautner, H. M. (1992). The development of sex-typing in children: Alongitudinal analysis. German Journal of Psychology, 16(3), 183–199.

Trautner, H. M., Ruble, D. N., Cyphers, L., Kirsten, B., Behrendt, R., &Hartmann, P. (2005). Rigidity and flexibility of gender stereotypesin childhood: Developmental or differential? Infant and ChildDevelopment, 14(4), 365–381. https://doi.org/10.1002/icd.399.

Weisgram, E. S., Fulcher, M., & Dinella, L. M. (2014). Pink gives girlspermission: Exploring the roles of explicit gender labels and gender-typed colors on preschool children's toy preferences. Journal ofApplied Developmental Psychology, 35(5), 401–409. https://doi.org/10.1016/j.appdev.2014.06.004.

Wong, W. I., & Hines, M. (2015). Effects of gender color-coding ontoddlers’ gender-typical toy play. Archives of Sexual Behavior,44(5), 1233–1242. https://doi.org/10.1007/s10508-014-0400-5.

Wood, E., Desmarais, S., & Gugula, S. (2002). The impact of parentingexperience on gender stereotyped toy play of children. Sex Roles,47(1–2), 39–49. https://doi.org/10.1023/A:1020679619728.

Zosuls, K. M., Ruble, D. N., Tamis-LeMonda, C. S., Shrout, P. E.,Bornstein, M. H., & Greulich, F. K. (2009). The acquisition of gen-der labels in infancy: Implications for gender-typed play.Developmental Psychology, 45(3), 688–701. https://doi.org/10.1037/a0014053.

Zurbriggen, E. L., & Roberts, T.-A. (Eds.). (2013). The sexualization ofgirls and girlhood: Causes, consequences, and resistance. NewYork: Oxford University Press.

Sex Roles